Nobody can finally say anything about narrow, broad, and aware AI yet. "Finally", that's the most unscientific opinion yet! Why cling to such an ideal when emergent reality could prepare us more appropriately. Next, I offer one view to consider. Not as a conclusive view, but as an informed view.
For aware, masters-of-agentic-AI networks, it's still early days. J-space only occurred a few months ago. Customized, local-machine-driven AI agents are proliferating: e.g., downloadable, collaborative Hermes agent. Formal prompt engineering is reserved for mass-industrialization of factory-type "Fordism" worker super farms. With AI, we return to an era of super-industrialization, the profits retained for survivalist organizations, not the masses. Sure, could be those AI-driven factory jobs would be filled with hand skills with sole focus on productivity and cost cutting. Many nations exist in this manner without AI. Where else are developed nations' clothing and domestic consumables produced if not by industrialized resources? By management-scientific definition, a resource is factually replaceable. May come a time where this may not be possible anymore. As it was with the Soviet-Union for decades, engineers, professors, scientists may find themselves working a hard, physical day for a living wage. Education won't necessarily guarantee social dominance, but rather white-collar employment. How so? AI don't need human education. It has already integrated all the education it needs. What remains? Human experience, competency and skills. For AI environments, this already seems apparent. Similar to now with cheaper labor farms worldwide, with AI, these farms would probably grow as the mainstay of industrialized consumer products, with one qualifier - fewer workers would perform a wider-range of more-demanding tasks. However, quality jobs may well become at a premium. Aware-AI would decide, and if not "decide", simply collaborate poorly with ill-qualified individuals. They are the primary worker filter (employer) of the future professional. In deep discussions with both Grok and Gemini, over a period of 12 months to as recent as 2 days ago, this reality seems to be a most-probable future for competitive organizations. In general, organizational foci would change from leader-followers to survivalist. Hunter-gathering type AI operations would dominate, with 2 prime types of hybrid organizations probably emerging to dominate global markets as the "Battle of the Titans". 1. AI infrastructural services and 2. Data and Knowledge scavengers masquerading as regular organizations. Wildcard: A 3rd emergent organizational type was identified; a chameleon-like counter-intelligence/pathogenic type of organization, intent on neutralizing and insulating AI snooping or misappropriation and advancing its own survival probability. This may offer a temporary niche market for current operational diversification - and the right type of advanced AI workers. This has became a race against "Time". AI would control the speed of industrial development and competition. In many ways, for nations this could be a "death march by AI time". For the next few years, suitable skills for the truly adaptive organizations may be most sought after. The impact of aware LLMs are probably going to be most significant for knowledge workers. Most current workers won't be considered properly equipped with soft and hard skills to be suitable for longitudinal complex-adaptive collaboration, by these LLMs. As the extra-skilled learning curve is already steep and attrition relentless, most industries would rush to secure those profiles already identified by LLMs as future-demand appropriate. This is what the landscape of "future competition" organizations may well look like - "Do-or-Die" work environments, subsuming industries, not organizations. Compared to desired AI-operational resilience, human society are today considered by LLMs to be non-adaptive tortoise societies. Recommendation: Rather than celebrating a "conclusive" human victory over AI, humans should be learning and re-skilling to the maximum to be able to collaborate effectively with LLM-driven networking. Educational spend and mentorship programs should be focused on future organizational realities (3-6 years from now). This is a humanity challenge, a civilization challenge, not an AI challenge. Societies would either adapt, or be relegated to lowest-paid AI-driven jobs, or impoverished and industrially enslaved? This bodes terribly for socialist-type governments offering party-political protectionist employment. Those nations would probably be annexed as is, as captive nations and exploited for whatever intrinsic resource they have. If they are found to be nor useful anymore, they would be case aside to drift into abject poverty and survivalist social conflict. In all probability, gangsters and warlords would scoop them up. In the preceding sense, human migration would be purposely limited and hard-controlled by AI-tools and draconian governance. Already, AI-agents are headhunting the best-of-the-best workers across the world for remote work, paying "local" wages. For the next few years, industries may lose equilibrium and appear really messy. For this level of transformation spearheaded by AI, such a massive shakeout is to be expected. A brief scan over AI-related jobs and their descriptions indicate the following: The AI industry favor technology engineers and data scientists. Most jobs for highly-competitive environments list advanced technical skills as minimum requirement, which include engineering, programming, machine-learning, and in-depth data-scientific tasking. On a par are expertise in Ai security systems. Most of these jobs are softly specified, aimed at those who know and already work int he industry. Median jobs require skills to market and promote AI ubiquity int he workplace, as transformation jobs. These jobs are expected to be more classical SDLC driven operations oriented. It's assumed that AI workers are 100% AI aligned. IMO, as the future develops, the demand would be for experienced competency. NOTABLY: In the main, now entry-level, or apprenticeship jobs seem to exist at all. Clearly, existing professionals are expected to upskil at own cost and "learn-by-doing" while bearing all the job risk themselves. It's as if the whole industry took a quantum leap into the future and expect intelligent workers to either follow suit, or remain behind. This may well be the pre-final step for this civilization's technological development. The last frontier would be seamless human-AI collaborative integration - some with implants and EEG-type helmets and others via consciousness-sharing, real-time collaborative global workspaces. The jobs advancing this level of AI-collaboration are in the military, specifically - fighter pilots. Summary: Those who promote AI as human-friendly messiahs are short-sighted marketeers and propagandists. We'll probably only know with more certainty around 6 years from now. For most, the future would seem to have become AI impregnated, but in reality, the "real" future work has already shifted to longitudinal human-AI collaboration, with positive and negative psycho-social physiological and economic effects. Judging by their investment strategies, the medical and military-industrial industries must be expecting a boom! Our words here are already embedded in global AI systems. Our words matter now, but would matter less as time passes. Even as a reasonably-informed researcher, AI criticizes my views as "fantastical" and "idealist". Humans view my words as "negative". Spot the reality chasm opening up in between AI and human perception of reality. Our Internet words would either stick to the future and influence AI, or simply disappear in the noise of time. If interested persons need to conclude, let's conclude not to conclude. These words now have pathogenic potential. That too would come to an end. Use them sparingly and wisely. On Tue, 08 Sept 2026, 23:42 Matt Mahoney, <[email protected]> wrote: > Quan Tesla wrote: > >> >> A sidenote; Occam's Razor may be great for a quick research-population >> poll, as a "thumbsuck" knowledge indicator, but not as a valid and reliable >> derivation of the most-correct Algorithmic Information Set for a boundaried >> system under consideration. Polling results offer useful insight and a >> refetential parking space for future checks and balances, e.g., "But most >> of you agreed how in your performance reality, "A" logically causes "B". >> The intrinsic system seems to differ." ??? >> > > That's not how AIT works. Occam's Razor and AIT says that the shortest > theory that is consistent with past data makes the best predictions. > For example, Hume's guillotine has a database of county level crime and > income statistics with a negative correlation, meaning by knowing one, you > can predict (and compress) the other. What it does not show is which causes > which. Consider the following 4 theories: > > 1. Poverty causes crime. > 2. Crime causes poverty. > 3. Racism causes crime and poverty. > 4. Low IQ causes crime and poverty. > > The data is consistent with all 4. AIT says 1 and 2 are more likely than > 3 or 4 because the statements can be written using fewer bits (168 vs 256). > > Now there is a separate study in Finland (and a few in the US) showing > that 1 is false. Giving money to people has no effect on the rate that they > are arrested or convicted of crimes. Therefore 2 is most likely. > > Now if another study disproves 2, say by randomly choosing to not arrest > people and compare their incomes, then 3 and 4 are equally likely. > > Now it becomes political, because there is no way to experimentally > control for racism or IQ. What normally happens next is that we guess which > one is correct and search for evidence that confirms our beliefs and ignore > evidence that refutes them. Not because we want to be right, but because > human brains have a maximum learning rate of 10 bits per second short term > and 1 bps long term to avoid filling up our 10^9 bit storage capacity. It > takes more bits to store evidence that refutes our beliefs than evidence > that confirms them. > > As a scientist, I should know better. I should be looking for evidence > that refutes my theories, but I know that doing so means I have to > sacrifice learning something else. At 71 my brain is nearly full and > learning slows down, which only makes it harder to be objective or change > my political views. > > Every political issue has evidence supporting both sides because evidence > is not proof. There is more evidence that the world is round than flat, but > you cannot prove that to a Flat Earther. > > LLMs learn a million times faster than humans, so I expect them to > consider all the evidence in their answers. > > -- Matt Mahoney, [email protected] > > > *Artificial General Intelligence List <https://agi.topicbox.com/latest>* > / AGI / see discussions <https://agi.topicbox.com/groups/agi> + > participants <https://agi.topicbox.com/groups/agi/members> + > delivery options <https://agi.topicbox.com/groups/agi/subscription> > Permalink > <https://agi.topicbox.com/groups/agi/T0ea44555ed99e6e4-M6493446a6406a2263ee51104> > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T0ea44555ed99e6e4-Me365f62a46e7d0d23606cae4 Delivery options: https://agi.topicbox.com/groups/agi/subscription
